Nagoya J. med. Sci. 35: 91-101, 1973

CLINICAL INVESTIGATION ON GLUCOSE TOLERANCE IN LIVER CIRRHOSIS

KATSUMOTO KATo, SABURO KIKUCHI, MADOKA ITo, SHIGERU HmANo,

TAKAHIKo 0YA, HIROSHI BANNo, AKro TaMURA,

ToMITSUGU KoYAMA, TAKEHIKO KoMATsu, TAKESHI TAKEI,

KorcHI ToMIMURA, AND MAsAHIRo YAMAUCHI

2nd Department of Internal Medicine, Nagoya University School of Medicine (Director: Associate Prof. Saburo Kikuchi)

ABSTRACT

Forty-six patients with liver cirrhosis were classified into three groups ac­cording to glucose tolerance test and the presence or absence of diabetic state. The results of xylitol tolerance test, galactose tolerance test, BSP retention, ICG disappearance rate and other routine liver function tests were compared between these three groups.

The results show that liver cirrhosis in patients with overt diabetes was less severe than that in non-diabetics. It is conceivable, therefore, that the glucose intolerance in most of the cirrhotic patients with overt diabetes is not due to the so called hepatogenous diabetes, but to essential diabetes with concomitant cir· rhosis.

Intravenous xylitol tolerance test may be available for the differential diagnosis of hepatogenous diabetes from essential diabetes with concomitant cirrhosis, since the change in serum glucose level after the administration of xylitol was characteristic in each group. Further studies must be done to substantiate this possibility.

INTRODUCTION

lt has been reported by many authors 1>2>3> that most patients with liver cirrhosis show abnormal carbohydrate tolerance. The mechanism of this carbohydrate intolerance is, however, far from clear, and many speculations as to its mechanism have been made by various authors. It is hard to dif­ferentiate the diabetic state resulting from liver cirrhosis itself from the essential diabetes with concomitant cirrhosis.

In our clinical investigation 4>5>, using xylitol and galactose tolerance tests, it was demonstrated that xylitol and galactose metabolism are disturbed in hepatic disorders and that the change of serum glucose level after the admin­istration of xylitol and galactose has a close relation to glucose intolerance.

1Jniliffl:k, *i.tt!r:=:!i!~. #l"Jii F:J, flEs:l!!tf ~. :kmlliS(i~. :!:B[!J!f !f,t;, J=l*flj)j:;t;, 'N-Li&li!PJ, ,J,:f.Rflol!;~. Itt#- ~. &H~-. W-!Ji3~5.L. Received for publication September 20, 1972.

91

92 K. KATO ET AL.

The present study was undertaken to investigate the abnormalities of

monosaccharide metabolism in liver cirrhosis and the possibility to differentiate

diabetic state due to liver cirrhosis from essential diabetes.

MATERIALS

Forty-six patients with liver cirrhosis were classified into three groups,

A, B and C as follows;

A: patients showing normal or borderline type in 50 g glucose tolerance

test (GTT) B: patients showing diabetic type in GTT, without fasting hyperglycemia

(higher than 120 mgjdl) and glucosuria 2 hours after meals

C: patients showing diabetic type in GTT with fasting hyperglycemia

and glucosuria 2 hours after meals

Glucose tolerance test was evaluated according to the report 6l of the

committee (Japan) on the diagnostic criteria of the oral glucose tolerance

test for diabetes mellitus. Forty of the forty-six patients were males and six were females. Their average age was fifty years (range 26-69). Liver cir­

rhosis was histologically confirmed by needle liver biopsy under laparoscopy

in thirty-eight patients and eight patients were diagnosed clinically without

histology.

Only five of the forty-six patients showed normal type in GTT and sixteen

patients showed borderline type. Twenty-one cases of the forty-six patients

with cirrhosis were classed as group A, ten as group B and fifteen as group

C. Table 1 shows that diabetic type in glucose tolerance test was more com­

mon in men than in women and that there were no significant differences in age among each group. Only one of fifteen cases (6.7%) had ascites in group

C, while five of twenty-one cases (23.8%) had ascites in group A and six of

ten cases (60%) had ascites in group B.

TABLE 1. Sex Ratio, Age-Structure, and Proportion of Ascites or the Treatment with Diabetogenic Drugs in Each Group

----·=====p=~=;====;====;==

Whole A B - -~-

M:F

Mean age (range)

No. with ascites

No. recieving g!ucocorticoids

No. recieving thiazide etc.

Total

group c

40 : 6 16 : 5 9 : 1 1 15 : o

1 12 1 s 1 6 1 1

I 2 I 2 I o I 0

I s I 4 I 3 I 1

1 46 1 21 10 · . I 15

GLUCOSE TOLERANCE IN LIVER CIRRHOSIS 93

Although two of the forty-six patients had been treated with glucocorticoid before this classification, these two patients were both classed as group A. In eight of the forty-six patients, oral diuretics of the thiazide or furosemide had been used. Four of the eight were classed as group A, three as group B and only one as group C.

METHODS

After an overnight fast, the following tolerance tests and various liver function tests including ICG or BSP test were performed.

(1) Oral xylitol tolerance test (0-XTT); 30 g of xylitol in warm water were given orally. Blood was then drawn every 30 min. for two hours.

(2) Intravenous xylitol tolerance test (IV-XTT) .: 20 ml of 50% xylitol solution were given intravenously. Blood was then drawn every 10 min. for 40 minutes.

(3) Galactose tolerance test (GaTT); 30 g of galactose in 300 ml of warm water were given orally. Blood was drawn every 30 min. for two hours.

Each serum was analyzed by gas-liquid chromatography for xylitol, gala­ctose and glucose according to the method of Sweely 7l. The results of these tolerance tests and other liver function tests in group A, B and C were compared with each other, or compared with the results in patients with chronic hepatitis, diabetics and healthy controls.

RESULTS

0-XTT Fig. 1- 1 shows the means of serum xylitol concentration at each interval.

In the healthy control group the serum xylitol level increased slightly (3.5 mgfdl) at 30 min. after the administration and then decreased gradually. The level was not zero even at 120 min. In the cirrhotic groups, as well as in chronic hepatitis, it rose to remarkably high levels at 30 min. and 60 min. The xylitol concentration course in diabetics was between that in healthy controls and that in liver disease patients. In order to estimate the xylitol concentration course with one figure, the sum of serum xylitol levels of each 30 min. for 2 hours was denoted as ZXyli. As shown in Fig. 2- 1 and Table 2, ZXyli. was quite high in the liver disease group compared with that in the diabetics and healthy controls. It was higher in group B than in group A and C, but the differences between each group were not statistically significant.

Fig. 1- 2 shows the change of serum glucose level after the oral administra­tion of xylitol. While there was almost no change in serum glucose level in groups A and B as well as in healthy control, there was a striking increase in diabetes and a moderate increase in group C. As shown in Fig. 2-2, Zt.Glu. (the sum of the differences between fasting serum glucose level and glucose

94 K. KATO ET AL.

mg/dl mg/dl

30 30

liver cirrhosis B

20 20

10 10

liver cirrhosis C

min.

1. Serum xylitol level course after the oral

administration of xylitol ( 30 g' ).

liver cirrhosis A

2. Change in serum glucose level from

zero-time after the oral tion of xylitol ( 30 g).

administra·

FIG. 1. Oral xylitol tolerance test (0-XTT).

mg/ dl

100

50

"' "' A B c ~

~;E 0

..,~

g~ ·~ "' = 0. .,-0 " ""'" - 0

liver ~-= .~ E "' " ..c: " cirrhosis " ""' ..c:

1. Sum of serum xylitollevels of each

30 minutes for 2 hours. ( .EXyli.)

mg/ dl

100

50

rh rh :~

Li-J "' 0

..,~ rn .,a ,.,~

·~ "' .s:.:: ..c:~

A B c = 0. .,- ~ = 0 " .o"' - 0 ~-= -~ ~ "'" ..c: " liver " ""'

..c:

cirrhosis Mean ± SD

2. Sum of increases of each 30 minutes

in serum glucose level from zero­

time for 2 hours. ( l'AGlu.)

FIG. 2 . .EXyli. and l'AGlu. (0-XTT).

GLUCOSE TOLERANCE IN LIVEL CIRRHOSIS

TABLE 2. Results of the Tolerance Tests Using Xylitol, Galactose, BSP and ICG in Each Group

95

I 0-XTT I IV-XTT ~~ GaTT I BSP4sl ICGK l'Xyli. l't.Glu. T 1/2 l't.Glu .. l'Gala. l't.Glu.

A (21 ) JMeanJ 41.6 -2.5 1 17.7 -19.6 1 415.8 29.3 122.9 10.091 SD 42.0 ( 15 )* 17.5 6.3 ( 15) 18.2 157.5 ( 12) 47.9 110.9(15) 0.048(17)

B <10) ~ ~~an~ ~~:! (6) 2~:~ ~ 2~:~ (6) 2~:~ I 2~~:~ (0), ~!:~ [ii:g (7)18:6g(8)

I' Mean [ 34.5 32.7114.8 46.3 I 212.4 149.4 115.4 [ 0.114 -C <15 l SD 14.2 (5) 40.1 3.5 (9) 24.3 128.8 (8) 61.5 7.2 (9)10.050(8)

hc::~mfs 1 ~~n 1 ~~:~ ( 6) i~:~ 11 ~:§ ( 9) - ~g:t 1 igi:~ ( 26) ~~:~ I I

~:r~1~~~ ~~~an i 1i:~ (11) 1~~:~ l li:~ (7) ~n I ~~:! (31) m:~ I I ~oe:i;~rs 1 ~~an 1 ~:~ ( 10) - 2~:~ [ 1i:~ ( 8) ~g:~ 1 ~u ( 17) - 1!:~ I I

*: number of cases

level at each interval) was significantly higher in group C than in healthy controls, and .E"Glu. in groups A and B was also higher than in healthy controls, but the difference was not significant. There were no significant differences in .s ... Glu. among the cirrhotic groups.

IV-XTT

As shown in Fig. 3-1, serum xylitol level decreased, showing exponential curve in every group. In liver disease groups, the decrease in serum xylitol level was remarkably delayed compared with that in healthy controls. Since the logarithm of the serum xylitol concentration decreased linearly with time in almost all cases, we estimated T 1/2 (time in minutes required for 50% decrease in serum xylitol level) as a suitable in IV-XTT. Fig. 4-1 shows that T 1/ 2 was significantly longer in cirrhotic groups than in diabetics or healthy controls and it was the longest in group B. There was a significant difference in T 1/2 between groups B and C.

Fig. 3-2 shows the change of serum glucose level after the intravenous administration of xylitol. The serum glucose level increased remarkably in diabetes and moderately in group C, while it remained almost constant in group B and decreased slightly in group A. .E,.,Glu. in diabetes was significantly higher than in any other group. Characteristically, .E.;Glu. was positive in group C, negative in group A and almost zero in group B. There were signigicant differences in .EAGlu. among the cirrhotic groups. (Fig. 4- 2)

96 K. KATO ET AL.

mg/dl mg/dl

40

B

20

10 20 30

40

liver cirrhosis C

chronic . hepatitis

40 min.

diabetes mellitus

1. Serum xylitol level course after the in· travenous administration of xylitol ( 10 g)

2. Change in serum glucose level from zero-time after the intravenous ad· ministration of xylitol ( 10 g)

FIG. 3. Intravenous xylitol tolerance test (IV·XTT).

min.

30

20

10

'" '" A B c -~·~ ,.a 0

'------.,----' !;::: ..,~

co. ""'~ ~ ~ ~" liver 0 ~ ~=

-o 1-<'..C: ~" cirrhosis ""' ;; ~

" ""'

1. Time in minutes required for de­crease of serum xylitol level to half. ( T 1/2)

mg/dl

100

50

A B C ~-~

liver ·cirrhosis

Mean± SD

2. Sum of increases of each 10 minutes in serum glucose level from zero time for 40 minutes. (L.'.~oGlu)

FIG. 4. T 1/2 and 2.:~.Glu. (IV-XTT).

GLUCOSE TOLERANCE IN LIVER CIRRHOSIS 97

mg/dl mg/ dl

!50

100

50

chron ic hepatitis

1. Serum galactose level course after the oral administration of galactose ( 30 g)

! 50

100

diabetes mellitus

healthy control

2. Change in serum glucose level from zero-time after the oral · administration of galactose ( 30 g )

FIG. 5. Galactose tolerance test (GaTT).

GaTT

Fig. 5-l shows the alteration of serum galactose level. In the healthy controls serum galactose level increased slightly at 30 min. after the admini­stration and almost zero at 90 min. In the cirrhotic groups it increased remarkably reaching the peak level at 60 min. The highest concentration was obtained in group A. As shown in Fig. 6- 1, .L:Gala. (the sum of serum galactose levels at every 30 min. for 2 hours) in cirrhotic groups was signi­ficantly higher than in diabetes or healthy control. .L:Gala. in group A was significantly higher than in groups B and C.

Fig. 5- 2 shows the change in serum glucose level after the administration of galactose. While serum glucose level remained constant in healthy controls, it increased remarkably in group C as well as in diabetic group and increased moderately in groups A and B. As shown in Fig. 6-2, 2:11Glu. in group C was significantly higher than in groups A and B.

Other routine liver function tests BSP retention values (% at 45 min.) in group C were lower than in groups

A and B, but the differences among the cirrhotic groups were not significant. ICG disappearance rate in group A was lower than in groups B and C. However, there were no significant differences among the cirrhotic group. (Fig. 7) As shown in Table 3, choline-esterase activity in group C was not

98 K. KATO ET AL.

mg/dl

400

200

A B C

liver cirrhosis_

1. Sum of serum galactose levels of each 30 minutes for 2 hours. (l'Gala.)

mg/dl

400

200

A B c .z "' 0

~ .S!-~ !~ >..b ~ "" .0: ~

liver 0 ~ v- .=; 0 ,~

cirrhosis '-"" -~ e " " "" ~

" '"" "" Mean± SD

2. Sum of increases of each 30 minutes in serum glucose level from zero-time for 2 hours. (l't.Glu.)

FIG. 6. 4Gala. and l't.GIU. (GaTT).

%

30

20

10

A

BSP retention

(%at 45 min. )

B c

ICG

disappearance rate

0.1

0.05

A B c Mean± SD

FIG. 7

GLUCOSE TOLERANCE IN LIVER CIRRHOSIS

TABLE 3. Results of Routine Liver Function Tests in Each Group

GPT (Unit)

Al. ph (K-K Unit)

ChE (L1pH)

Albumin (g/dl)

r-globulin (g/ dl)

A/G I

TTT (Maclagan U.) I cholesterol (mg/ dl) I

Mt;an ± SD

A (21) B (10) c (15)

78 ± 65 (21)* 1 36 ± 23 (10) 54 ±15 (15)

9.3 ± 5.9 (18 ) 1 1o.s ± 7.1 (9) 11.8 ± 4.1 (15)

0.48 ± 0.20 (19) I 0.37± 0.14 ( 8 ) I 0.60±0.1S (11)

3.5 ± 0.9 (20) I 3.5 ± 0.6 -~IV ±0.6 (12)

2.1 ± o.s (18) 1 1.8 ± o.6 (9) 1 1.9 ± o.6 (12)

uo± o.37 (20) 1 1.02± o.26 (10) 1 1.06± o.3o (14)

8.3 ± 3.2 < 191 1 6.2 ± 3.2 < 9) 1 s.s ± 3.6 < 14)

160 ± 35 (18) 1151 ± 25 (8) 1 180 ± 41 ( 14)

*: number of cases

99

so low as those in groups A and B, and the difference between groups B and C was significant statistically.

DISCUSSION

The data obtained in this study confirm the high incidence of glucose intolerance in patients with cirrhosis in accord with previous reports. In the present study, fifteen of forty-six patients had not only diabetic glucose tolerance curve, but also such diabetic signs as fasting hyperglycemia and glucosuria at 2 hours after meals (group C). The cause of glucose ittreler-anee-i-ft-i)atients­with cirrhosis is still unknown. Probably it may be due to a combination of several mechanisms. Creuzfeldt et at.8>9> proposed the conception of "hepato­genous diabetes" that chronic parenchymal liver disease leads to impairment of glucose tolerance and eventually to clinical diabetes. Megyesi et al.1>

proposed that glucose intolerance in chronic liver disease_ is directly or indirectly a consequence of insulin resistance. On the other hand, some of patients with essential diabetes may be associated with concomitant cirrhosis. However, in any individual subject we cannot differentiate clearly the diabetic state attributable to liver cirrhosis from the essential diabetes with concomitant cirrhosis. The patients in group C in this study, therefore, are composed of both the so called hepatogenic diabetes cases and the essential diabetes cases with concomitant cirrhosis.

The liver function tests and clinical signs in group C were, therefore, compared with those in groups A and group B, in order to investigate the mechanism of diabetic state in liver cirrhosis. Although some patients were treated with potentially diabetogenic drugs such as glucocorticoids, thiazide

100 K. KATO ET AL.

or furosemide, only one of these patients were classified into group C. Age distribution was not different in all three groups.

From the results of xylitol and galactose tolerance tests, it was evident that each three group showed a characteristic tolerance curve which was evaluated statistically. In GaTT, galactose was rather easily metabolized in group C compared with groups A and B. In 0-XTT and IV-XTT, xylitol in the blood of group C was eliminated more quickly than in groups A and B. From the clinical and experimental investigations reported previously by us4' 5>, galactose and xylitol tolerance tests have been proven to reflect well the degree of liver damage and to be valuable liver function tests. So, the results of these tolerance tests show that there was not so severe liver damage in group C as in groups A and B.

As for BSP and ICG tests, the retention in groups A and B was more remakable than in group C. Choline-esterase activity in group C was not so low as in groups A and B. Moreover, diabetic state is less common in cirrhotic patients with ascites. These results also show that there was no severe liver damage in group C.

In so called hepatogenous diabetes, if it exists, it is anticipated that the occurrence of diabetic state has good correlation with hepatic disorder and also with age, but our results did not accord with this anticipation. So, it is conceivable that the diabetic state in most of the cirrhotic patients in group C is not of hepatogenous origin but due to essential diabetes with concomitant cirrhosis. All the cases where the hepatic disorder precedes the diabetic state are not hepatogenous diabetes, because it is possible that the latent diabetes advances to be overt owing to hepatic disorder in some cases.

The changes in serum glucose level in IV-XTT were characteristic in each cirrhotic group. Serum glucose level remained almost constant in group B, while it rose in group C and decreased slightly in group A. Consequently, it seems to be quite useful to apply this characteristic result for differentiating group B from group A and group C. The cases in group B did not show so remarkable glucose intolerance as those in group C. As for the results of many liver function tests including 0 -XTT, IV-XTT and GaTT, the former showed more severe liver disorder than the latter. Therefore, most of the cases in group B may be considered to have mild glucose intolerance due to liver cirrhosis itself and to correspond to the forestage of hepatogenous diabetes, although a few cases may show slightly abnormal glucose tolerance due to the concomitant essential latent diabetes. On the other hand, the diabetic state in most of the cases in group C is considered to be due to essential diabetes with concomitant cirrhosis. From these evidences, it may be allowed to provide a working hypothesis that hepatogenous diabetes shows almost constant or slight increase in glucose level after lV-XTT and essential diabetes with concomitant cirrhosis shows moderate or remarkable increase.

GLUCOSE TOLERANCE IN LIVER CIRRHOSIS 101

This hypothesis may be possibly available for differential diagnosis of hepato­genous diabetes and essential diabetes with concomitant cirrhosis in group C. In order to substantiate this possibility, further studies must be done, including the clinical long-term observation of many cases, especially of cases progres­sing from group B to group C. Serum glucose level in 0-XTT and GaTT also showed as similar changes as that in IV-XTT, but these changes were not verified to be significant statistically.

CoNCLUSION

In an attempt to differentiate hopatogenous diabetes from essential diabetes with concomitant cirrhosis, forty-six patients with liver cirrhosis were investi­gated, using xylitol and galactose tolerance tests.

1. It may be concluded that the glucose intolerance in most of the cirrhotic patients with overt diabetes is due to essential diabetes with concomitant cirrhosis and that the so called hepatogenous diabetes is not common in cir­rhotic patients.

2. Intravenous xylitol tolerance test may be available for the differential diagnosis of hepatogenous diabetes from essential diabetes with concomitant cirrhosis.

REFERENCES

1) Megyesi, C. et al., Glucose tolerance and diabetes in chronic liver disease, Lancet, 1051, 1967.

2) Carneiro de Moura et al., Carbohydrate metabolism studies in cirrhosis of the liver, Amer. f. Med., 13, 891, 1968.

3) Conn, H. 0. et al., Cirrhosis and diabetes I, Increased incidence of diabetes in patients with Laennec's cirrhosis, Amer. ]. Med., 14, 837, 1969.

4) K. Kato et al., Clinical and experimental studies on xylitol metabolism, ]ap. ]. Med., 11, 92, 1972.

5) S. Kikuchi et al., Studies on galactose tolerance test in patients with liver disease, Nagoya ]. med. Sci., 35, 19, 1972.

6) N. Kuzuya et al., Report of the committee on the diagnostic criteria of the oral glucose tolerance test for diabetes mellitus Recommendations on the evaluation of the oral glucose tolerance test for the diagnosis of diabetes mellitus, ]. ]ap. Diab. Soc., 13, 1, 1970.

7) Sweely, C. C. et al., Gas-Liquid chromatography of trimethylsilyl derivatives of sugars and related substance, ]. Am. Chern. Soc., 85, 249/', 1963.

8) Creutzfeldt, W. et al., IntravenBse belastungen mit glucose, insulin und tolbutamide bei gesunden, diabetikern, leberzirrhotikern und insulomtriigern, Deutsch Med. Wschr., 87, 2189, 1962.

9) Leevy, C. et al., Hyperglycemia and glucosuria in the chronic alcholic with hepatic

insufficiency. Clinical obserbations in 10 patients., Amer. J. Med. Sci., 223, 88, 1961.